bug-investigator

Reproduce software defects and generate failing tests for root-cause analysis.

15|2|Updated Jan 8, 2026
One-click install
npx skills add https://github.com/GrishaAngelovGH/gemini-cli-agent-skills --skill bug-investigator
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bug-investigator
Source: https://github.com/GrishaAngelovGH/gemini-cli-agent-skills/tree/main/bug-investigator
Command: npx skills add https://github.com/GrishaAngelovGH/gemini-cli-agent-skills --skill bug-investigator

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill guides a systematic transition from vague symptom to verified fix by instructing on information gathering, reliable reproduction, focused root-cause analysis, and regression-safe remediation.

Core Features & Use Cases

  • Symptom Analysis & Scoping: Determine what is failing, where it occurs, and which components or files are implicated.
  • Reproduction & Test Automation: Create minimal reproducible cases and automate failing unit or integration tests to validate the issue.
  • Root Cause Analysis & Verification: Use a checklist-driven 5-whys approach, trace data and state, review recent changes, apply minimal fixes, and verify by running targeted and full test suites.
  • Prevention: Add guardrails such as assertions, type checks, and improved logging, and document the rationale for non-obvious fixes.
  • Use Case: Ideal for intermittent crashes, regressions introduced by recent commits, failing CI tests, or unexpected runtime behavior across frontend and backend code.

Quick Start

Investigate the reported bug by reproducing the failure, creating a minimal failing test, running checklist-driven root-cause analysis, and proposing a minimal verified fix.

Frequently Asked Questions about bug-investigator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find the root cause of an intermittent software crash?

Find the root cause of an intermittent crash by performing systematic bug investigation: gather failure symptoms, create minimal reproducible test cases, and apply a checklist-guided root-cause analysis to trace data and state to the exact failure point.

What is the best way to reproduce a failing CI test?

The best way to reproduce a failing CI test is to build a minimal failing case that automates the issue across frontend and backend codebases. This validates the defect and confirms the runtime behavior before attempting any fixes.

How do I fix a regression introduced by a recent commit?

Fix a recent commit regression by reviewing recent changes, applying a minimal fix to the implicated components, and verifying the remediation by running targeted and full test suites to ensure regression-safe behavior.

Can I use root-cause analysis for both frontend and backend runtime failures?

Yes, you can apply root-cause analysis to unit, integration, and runtime failures across both frontend and backend codebases. The process requires reproducible test cases, targeted tracing, and logging to verify the affected test suites.

How do I prevent unexpected runtime behavior from recurring after a fix?

Prevent unexpected runtime behavior from recurring by adding guardrails such as assertions, type checks, and improved logging. Document the rationale for non-obvious fixes to maintain regression-safe remediation and clarify future debugging.